I am trying to replicate Robert Barro's results in his 1991 Economic Growth in a Cross Section of Countries paper. I have had problems with the deviation of PPP of investment. I don't know if I am just supposed to obtain the plain difference between the country's value and the sample mean or if this involves more extensive calculation.
Statistics: To replicate a result I need 2 get the observations' deviation from the mean. How do I do it?
The deviation of a value from the mean is the value minus the mean usually denoted as (x - xbar). However if you calculate all the deviations and add them together you will get zero, to avoid this, convention has it that we take the absolute value of each deviation, i.e. | x - xbar
Reply:|Its simple your over thinking it.
Deviation from mean = value - mean.
Its that simple.
Reply:the standard deviation has a formula sqrt n x p x q
the mean is n x p
n = no in sample that the mean measures
p = pr of the desired event happening
q = 1-p (ie event NOT happening)
Reply:we are talking statisics here, take the mean.
all stats are based on mean averages, do you know of any other?
remember the maxim
81% OF All statistics ARE MADE UP ON THE SPOT.
Reply:It might mean "how many standard deviations is it from the mean" -- its z-score.
That means, you take the standard deviation, and you take the deviation of the observation. How many (or what percent) of the first is the second?
Say the standard deviation is .5, and the deviation of the observation is 1. It's 2 standard deviations. If the observation is .25, it's .5 standard deviations from the mean.
As I say, I can't be sure this is what you're looking for.
my cat
No comments:
Post a Comment